Google model product
Gemini 4 Neon is Google DeepMind’s Gemini 4 reasoning model for software engineering, knowledge work, and science workflows.
Updated Nov 22, 2026. Default version: Gemini 4 Neon
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Gemini 4 Neon.
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkSEC-bench Pro | Score90.50% | Rank01 | Participants11 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkVending-Bench 2 | Score16,843.00 usd | Rank01 | Participants7 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Leaderboard | Score18.41% | Rank01 | Participants50 | Percentile100.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Praise Complaint | Score44.47% | Rank01 | Participants50 | Percentile100.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Steerability | Score21.20% | Rank01 | Participants50 | Percentile100.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Task Outcome Explicit | Score19.20% | Rank01 | Participants50 | Percentile100.00% | EvidenceA | Evaluated |
| BenchmarkChartography (no tools) | Score73.50% | Rank01 | Participants7 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkRiemannBench | Score84.00% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkVals Index | Score70.90% | Rank01 | Participants4 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkAgents' Last Exam | Score75.80% | Rank02 | Participants33 | Percentile96.88% | EvidenceC | Evaluated |
| BenchmarkGraphwalks BFS >128k | Score95.20% | Rank02 | Participants14 | Percentile92.31% | EvidenceC | Evaluated |
| BenchmarkLVBench | Score89.00% | Rank02 | Participants33 | Percentile96.88% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,509.28 rating | Rank02 | Participants135 | Percentile99.25% | EvidenceA | Evaluated |
| BenchmarkFrontierCode 2.0 | Score46.20% | Rank02 | Participants7 | Percentile83.33% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text | Score1,523.55 rating | Rank03 | Participants222 | Percentile99.10% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,512.94 rating | Rank03 | Participants222 | Percentile99.10% | EvidenceA | Evaluated |
| BenchmarkAutomationBench | Score56.30% | Rank04 | Participants33 | Percentile90.63% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench-Science 0.1 | Score70.50% | Rank04 | Participants17 | Percentile81.25% | EvidenceC | Evaluated |
| BenchmarkLM Arena Webdev | Score1,764.89 rating | Rank04 | Participants112 | Percentile97.30% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Tool Hallucination | Score0.99% | Rank18 | Participants50 | Percentile65.31% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Bash Reliability | Score-5.97% | Rank39 | Participants50 | Percentile22.45% | EvidenceA | Evaluated |
Preference and agent-evaluation results for the default version.
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenaagent | Categoryoverall | Rank01 | Rating / score0.18 | VotesN/A | Observations903.8K | Result date |
| Arenaagent praise complaint | Categoryoverall | Rank01 | Rating / score0.44 | VotesN/A | Observations1.2K | Result date |
| Arenaagent steerability | Categoryoverall | Rank01 | Rating / score0.21 | VotesN/A | Observations25.8K | Result date |
| Arenaagent task outcome explicit | Categoryoverall | Rank01 | Rating / score0.19 | VotesN/A | Observations2.8K | Result date |
| Arenatext | Categoryfrench | Rank01 | Rating / score1,562.86 | Votes183 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry life and physical and social science | Rank01 | Rating / score1,549.27 | Votes827 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank01 | Rating / score1,566.38 | Votes247 | ObservationsN/A | Result date |
| Arenatext | Categorymath | Rank01 | Rating / score1,547.63 | Votes212 | ObservationsN/A | Result date |
| Arenatext factuality | Categorycoding | Rank01 | Rating / score1,566.99 | Votes1,236 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry life and physical and social science | Rank01 | Rating / score1,551.32 | Votes710 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryrussian | Rank01 | Rating / score1,534.59 | Votes846 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank01 | Rating / score1,557.51 | Votes183 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank01 | Rating / score1,553.13 | Votes247 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev-html | Rank01 | Rating / score1,802.83 | Votes400 | ObservationsN/A | Result date |
| Arenatext | Categorycoding | Rank02 | Rating / score1,555.25 | Votes1,288 | ObservationsN/A | Result date |
| Arenatext | Categoryexpert | Rank02 | Rating / score1,559.70 | Votes423 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts | Rank02 | Rating / score1,547.08 | Votes3,508 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry software and it services | Rank02 | Rating / score1,541.64 | Votes1,944 | ObservationsN/A | Result date |
| Arenatext | Categoryinstruction following | Rank02 | Rating / score1,535.28 | Votes1,918 | ObservationsN/A | Result date |
| Arenatext | Categorynon english | Rank02 | Rating / score1,524.13 | Votes3,701 | ObservationsN/A | Result date |
| Arenatext | Categoryrussian | Rank02 | Rating / score1,553.96 | Votes970 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryexclude ties | Rank02 | Rating / score1,529.24 | Votes4,179 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryhard prompts | Rank02 | Rating / score1,543.07 | Votes3,457 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry software and it services | Rank02 | Rating / score1,544.66 | Votes1,905 | ObservationsN/A | Result date |
| Arenatext factuality | Categorymulti turn | Rank02 | Rating / score1,518.61 | Votes688 | ObservationsN/A | Result date |
| Arenatext factuality | Categorynon english | Rank02 | Rating / score1,505.71 | Votes3,652 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryoverall | Rank02 | Rating / score1,509.28 | Votes5,486 | ObservationsN/A | Result date |
| Arenatext style control | Categorycoding | Rank02 | Rating / score1,558.99 | Votes1,288 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry life and physical and social science | Rank02 | Rating / score1,535.52 | Votes827 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry medicine and healthcare | Rank02 | Rating / score1,527.67 | Votes323 | ObservationsN/A | Result date |
Official vendor API pricing appears first, followed by individual provider offers.
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderVertex | Provider model IDgemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderOpenCode | Provider model IDgemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderOpenRouter | Provider model IDgoogle/gemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderGitHub Copilot | Provider model IDgemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderGoogle | Provider model IDgemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderDevPass (LLM Gateway) | Provider model IDgemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderCrossModel | Provider model IDgemini/gemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderKilo Gateway | Provider model IDgoogle/gemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderEden AI | Provider model IDgoogle/gemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderVercel AI Gateway | Provider model IDgoogle/gemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderPrime Inference | Provider model IDgoogle/gemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderMerge Gateway | Provider model IDgoogle/gemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderRequesty | Provider model IDgemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderOfox | Provider model IDgoogle/gemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderNanoGPT | Provider model IDgoogle/gemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context983K | Updated |
| ProviderVivgrid | Provider model IDgemini-4-neon | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderCortecs | Provider model IDgemini-4-neon | Regionglobal | Input / 1M$3.41 | Output / 1M$17.06 | Context1M | Updated |
| ProviderVenice AI | Provider model IDgemini-4-neon | Regionglobal | Input / 1M$3.75 | Output / 1M$18.75 | Context1M | Updated |
Provider-specific output speed and catalog latency for Gemini 4 Neon. Runtime does not affect the capability score.
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Technical details for the model's default version.
Available versions of this model. The score column identifies the version used in the overall ranking.
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGemini 4 Neon | Released | LLMBoard96.24 | ParametersN/A | Context1M | Max output1M | Open weightsNo | LicenseProprietary |
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Key information about Gemini 4 Neon and its available data.
Gemini 4 Neon is a large language model from Google with a 1,048,576-token unified context window, text output, and support for text, image, video, audio, and PDF input. It offers low, medium, high, and xhigh thinking levels, with high as the default, and is available on the Gemini API paid tier.
Data as of 2026-11-21.
Common questions about Gemini 4 Neon.
Gemini 4 Neon's default version was released on Nov 18, 2026.
Gemini 4 Neon's official API price is $3 per million input tokens and $15 per million output tokens via Google. The lowest tracked third-party offer starts at $3 input and $15 output via OpenCode.
Gemini 4 Neon was created by Google.
The default version has a 1M token context window.
No. The default version is not marked as having publicly available weights.
18 provider offerings are linked to the default version.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.